How does speech-recognition software compare to an expert lip-reader in accuracy?

BillieBob

Clearly the lip-reader has the disadvantage of needing to see the speaker speak.
In a noisy restaurant however it is the computer that would be unable to translate.
So given a lip-reader that can see the speaker in a room together with a speech-
recognition system which one is more accurate?
Has the hearing-impaired community embraced this new technology and if so do they carry a portable system with them?

Answer
Technology has come a long way. EX my phone's voice recognition software is superb and cuts out background noise. I was blaring rammstein and spoke a text message to my girlfriend and it worked. Its a Motorola Droid btw.

What is the best GPS system?

casper32

With turn by turn directions with street names!!!!

Answer
Garmins are GREAT GPS systems so go with a Garmin Nuvi! From there it all depends on what features you want. General rule of thumbs is the higher the model number the more features and the more it cost. You can get a simple GPS that will get you where you need to go for about $130 or you can break bank with a high tech GPS towards $600 all depends on what features you want. Whatever model you choose to get I recommend getting the one that ends in '5' this is a newer model so the processor is faster, it finds a signal faster, and my favorite it shows you your current speed limit along with the speed limit of the road you are traveling on.

All Nuvi's will get you where you need to go even the lowest model will get you there the same way the more expensive ones would the differences are the costly features such as

*Text to speech is the ability to speak street names. A GPS will usually say something like "turn right in 3.2 miles" and then display the street name in writing but with text to speech the GPS will say "turn right in 3.2 miles onto Jackson Rd."
*Traffic- ability to see traffic jams and plan a route to avoid them. This is a yearly or monthly extra fee or free with Nuvi's that end with 'T'.
*Voice recognition. The ability to speak addresses into the Garmin without entering them in by hand.
*Bluetooth you can connect your phone to your GPS wirelessly if your phone allows it so you can place or answer calls on the GPS without the need to stumble around for your phone.
*MP3- can play mp3 through the GPS
*FM Transmitter- allows you to hear the GPS through your car speakers.

It sounds like you want a GPS that has text to speech so I would recommend the Garmin Nuvi 260. You can check out the reviews here, also the link has a chart called "Which nüvi is Best for You" which can help you find a GPS based on the features you want.
http://www.amazon.com/gp/product/B000UU7ZPS?ie=UTF8&tag=marartpla-20&linkCode=xm2&camp=1789&creativeASIN=B000UU7ZPS

Powered by Yahoo! Answers

Speaker Matching criterion (voice recognition related)?

Leonardo

Hi folks,

I'm not a native English speaker but I work as a localizer and translator at a company that develops voice recording and recognition software.
Recently, there has been a good deal of debate about one of the phrases in the GUI of one of our software solutions, if it was rendered into English correctly or not.
The case is that we have a fetaure that analyzes two recordings and can tell you if the speaker on recording 1 and 2 is the same or whether they are two different persons. Now, there is a basic criterion that allows software to tell you if the speaker is the same person or if there are two of them. The questions is how to properly name this criterion in English? Would "Speaker Matching Criterion" do? The disputable word here is Matching, but if Criterion is not good and there's a better option, please mention that, too.

Many thanks in advance.
Hey Bander,

Thanks for your response.
Well, the key point is not recognizing allophones but making a conclusion if the speaker is the same on both recordings or if they are different. The software makes decision itself, although the "listener" who's working with the software may change the decision manually.
When "coining" the "matching" term I was thinking about an identic situation and remembered we are given "passwords do not match" message when password and password confirmation are different. The similiarity lies in there being two objects that may be the same or different, and in the example above "match" is used. Is our case totally different then?
Btw, I also googled it and found there is "voice matching software" term used for the same thing. Maybe Voice Matching Criteria will be a good option then?

Thank you!

Answer
Yes,"matching" may not be a good choice. It seems that the system you developed can regognize the different sounds of a consonant (allophones), and interpret them as identical (matching). In this special apparatus, the process would be "categorical perception", similar to human perception of speach sounds. And since recognition of sounds here is eventually interpreted by the listener (not the speaker), you should maybe call it "speech categorical perception (apparatus)".

Edit: On the allophones and our peculiar ways of articulating sounds, I said it's impossible to have a match if the apparatus was producing spectographs of the speech signals. We are not machines; however, we can regognize the speaker's voice regardless of how sounds are produced. That's why I thought of "categorical perception". Your system does that, but I don't know how.

give me details of voice security system?

pandu

Answer
Suggest you read http://www.biometrics.gov/docs/speakerrec.pdf

or search on speaker recognition


NIST conducted yearly evaluations of speaker recognition system
http://www.nist.gov/speech/tests/spk/index.htm

Powered by Yahoo! Answers

Speaker Matching criterion (voice recognition related)?

Leonardo

Hi folks,

I'm not a native English speaker but I work as a localizer and translator at a company that develops voice recording and recognition software.
Recently, there has been a good deal of debate about one of the phrases in the GUI of one of our software solutions, if it was rendered into English correctly or not.
The case is that we have a fetaure that analyzes two recordings and can tell you if the speaker on recording 1 and 2 is the same or whether they are two different persons. Now, there is a basic criterion that allows software to tell you if the speaker is the same person or if there are two of them. The questions is how to properly name this criterion in English? Would "Speaker Matching Criterion" do? The disputable word here is Matching, but if Criterion is not good and there's a better option, please mention that, too.

Many thanks in advance.
Hey Bander,

Thanks for your response.
Well, the key point is not recognizing allophones but making a conclusion if the speaker is the same on both recordings or if they are different. The software makes decision itself, although the "listener" who's working with the software may change the decision manually.
When "coining" the "matching" term I was thinking about an identic situation and remembered we are given "passwords do not match" message when password and password confirmation are different. The similiarity lies in there being two objects that may be the same or different, and in the example above "match" is used. Is our case totally different then?
Btw, I also googled it and found there is "voice matching software" term used for the same thing. Maybe Voice Matching Criteria will be a good option then?

Thank you!

Answer
Yes,"matching" may not be a good choice. It seems that the system you developed can regognize the different sounds of a consonant (allophones), and interpret them as identical (matching). In this special apparatus, the process would be "categorical perception", similar to human perception of speach sounds. And since recognition of sounds here is eventually interpreted by the listener (not the speaker), you should maybe call it "speech categorical perception (apparatus)".

Edit: On the allophones and our peculiar ways of articulating sounds, I said it's impossible to have a match if the apparatus was producing spectographs of the speech signals. We are not machines; however, we can regognize the speaker's voice regardless of how sounds are produced. That's why I thought of "categorical perception". Your system does that, but I don't know how.

give me details of voice security system?

pandu

Answer
Suggest you read http://www.biometrics.gov/docs/speakerrec.pdf

or search on speaker recognition


NIST conducted yearly evaluations of speaker recognition system
http://www.nist.gov/speech/tests/spk/index.htm

Powered by Yahoo! Answers

How does speech-recognition software compare to an expert lip-reader in accuracy?

speaker recognition system image

BillieBob

Clearly the lip-reader has the disadvantage of needing to see the speaker speak.
In a noisy restaurant however it is the computer that would be unable to translate.
So given a lip-reader that can see the speaker in a room together with a speech-
recognition system which one is more accurate?
Has the hearing-impaired community embraced this new technology and if so do they carry a portable system with them?

Answer
Technology has come a long way. EX my phone's voice recognition software is superb and cuts out background noise. I was blaring rammstein and spoke a text message to my girlfriend and it worked. Its a Motorola Droid btw.

can a voting system which uses text dependent speaker recognition system be developed?

datta

what will be the cost of speaker recognition chips?

Answer
It is possible to do this, as some major communications companies (ie. Bell, Rogers, etc.) use voice recognition software that recognizes multiple voices, and can differentiate different types of answers.

Although it's not yet perfect, and some people still have to be put through to a representative, it does work to a certain extent. Therefore, it could be adapted to a voting system for your purposes.

Powered by Yahoo! Answers

How do British people spell their postcodes?

speaker recognition system image

kaheasmus

Hey everybody, of course this will be a simple question for native speakers. It results of a job i have to do at the moment which deals with speech recognition for different kinds of embedded systems. I just want to know how You spell the numbers in the postcodes(as single digits, connected or otherwise)
Thanks a lot.

Answer
one two three four five six seven eight nine

What is the best way to bundle audio cables?

Carefree A

Right now, the cables behind my audio receiver are tangled beyond recognition. I need to organize them...

How should I bundle my speaker wire to minimize RF and EM interference? I've heard that a "8-shaped" bundle is good.

Is there any truth to this claim? If so, explain.

Thanks in advance!

Answer
Keep power cables separate from Line Level cables(inputs, & record outputs)
and speaker cables away from Line Level cables.
In other words keep the different types of cables away from each other.
This also goes for FM antennas> separate them all as well as possible.
If you have HUM, you most likely have a cable that has a compromised (bad or going bad) Ground or loose connector.
Bundling wires, even the same kind CAN induce hum, as it turnes it cable bundle into a coil (like a Guitar Pick-up).
Try buying several extra good quality cables of each type, and switch them out one at a time, and you may find that one or more cable is compromised and is causing the trouble> and throw it away!
Make sure that the outlet the system is plugged into has a Good quality Surge Protector!> not just an outlet strip<
Make sure the circuit that is being used to power the system is not also carrying lights with dimmers, florescent lights, or any thing other than sound equipment. >LIGHTS & MOTORS CAUSE HUM!
Also make sure that your not using several circuits to power several INTERCONNECTED items!
>ie: VCR, DVD, CD, PHONO on one circuit, and Receiver, TV, Cable Box on another (GROUND LOOP).
Many times a moderately expensive Surge Protector will solve hum in a sound system(not a $10.00 one, but maybe $25.00 and up)
I use a TRIPPLITE that cost more than $200.00,
but it has been well worth it! ( I have a $50,000.00 studio
sound system), as it not only filters noise that comes
down the line, but it protects from surges & spikes.
(Duke Power calls them "LINE VARIATIONS"), but they will burn up delicate electronics in a heartbeat!
If your wires are bundled, it will be hard to clean up the mess and noise!
Even sharp bends in wires and cables will degrade the performance of even the best system, and make change outs (new DVD player) hatefull as well as put extra strain on the rest of the wires. and they wonder why they failed.

Powered by Yahoo! Answers